Search

Summary Ephemeral gullies or rills are major sediment sources and the primary channel to convey runoff and sediments from hillslopes. Quantifying the detachment rate is important for understanding the erosion process of ephemeral gullies. This study suggests a method to estimate the detachment rate for use in WEPP model, using sediment distribution data along an eroding rill or ephemeral gully under controlled flume experiments, conducted at four slope gradients (8.74%, 17.62%, 26.78%, and 36.38%) and four flow rates (32, 64, 128 and 256 L min −1 ). The experimental soil materials were packed into the flume of 0.5 m wide, 0.6 m deep and 12 m long, to a thickness of 0.5 m at a bulk density of 1300 kg m −3 . Regulated water flow was introduced to soil surface at the upper end of flumes. The detachment rate was calculated from the spatial distribution data of sediment concentrations measured with sediment-laden water samples simultaneously taken along the gullies at one meter intervals. Results showed that the detachment rate was well fitted with downslope distance along the rill or ephemeral gully by an exponential equation. The regression parameters indicated that detachment rate decreases linearly with sediment load in the gully flow. The feedback relationship between sediment load and detachment rate in rill or ephemeral gullies could be adequately represented by the first-order coupling equation for loess soil under steep slopes and high flow rates. The detachment rates along the rills or ephemeral gullies were found to increase with slope gradient and flow rate and the detachment rate was positively related to unit stream power and the deficit between max sediment load and the local sediment load. [ABSTRACT FROM AUTHOR]

Accurately describing the soil erosion process is crucial for improving the accuracy of existing simulation models of ephemeral gully erosion. However, a laboratory experimental strategy for simulating sediment distribution along slope lengths of an ephemeral gully is lacking. This limitation greatly hinders the development of simulation models for ephemeral gully erosion. In this study, we constructed six soil flumes with 12 m in length, 0.1 m in width, and 0.5 m in depth, which were filled with cultivated loess soils. Four flow rates (5.33, 10.67, 21.33, and 42.67 × 10−3 m3 s−1 m−1), four slope gradients (5°, 10°, 15°, and 20°), and six gully lengths (2, 4, 6, 8, 10 and 12 m) were adopted to perform a series of laboratory scouring experiments. The measured sediment concentrations at the outlet ranged from 23.39 kg m−3 to 534.60 kg m−3. The trend of exponential increase in sediment concentrations along slope lengths of an ephemeral gully was similar to that observed in a rill. According to the fitting equation, the maximum sediment concentrations (A e) ranged from 358.05 kg m−3 to 910.88 kg m−3. The attenuation coefficients (B e) indicated the rate of increase in sediment load with slope length, which ranged from 0.042 m−1 to 0.185 m−1. By combining the measured sediment loads along the slope length (8 m) with the sediment transport capacity, the sediment distribution along slope lengths of an ephemeral gully could be accurately quantified. Moreover, the results confirmed that a flume with a short slope length (4/6 m) could effectively determine the sediment distribution along slope lengths at gentle slope of 5° and 10°. However, a minimum length of 8 m was advisable at steep slope of 15° and 20°. In summary, our experimental strategy provides an efficient method for determining the sediment distribution along slope lengths of an ephemeral gully. [ABSTRACT FROM AUTHOR]